This month, Bayesian health became the first company to receive FDA clearance for an AI-powered continuous sepsis monitoring system.
Many AI tools for sepsis detection have historically operated without FDA review, being deployed as generalized clinical decision support software. Suchi Saria, founder and CEO of Bayesian, called the FDA clearance much bigger than a regulatory step: She sees it as validation of her startup’s broader vision for AI-powered “real-time clinical intelligence” that helps hospitals move from reactive to proactive care.
Saria noted that Bayesian has spent years working with the FDA to establish reliable definitions of sepsis, validate its model’s performance in various hospital settings and patient populations, assess risks such as missed cases and alert fatigue, and create a post-market monitoring and quality assurance program.
In his view, AI in healthcare requires far more rigor than most tools currently receive. She said any AI tool affecting patient care should receive this level of scrutiny, even if it doesn’t actually go through the regulatory process.
“A lot of people wrongly view FDA clearance as a ceiling, but I think it’s just a floor. That’s the starting point,” Saria explained. “If you can’t change the clinician’s action, you won’t get results – and changing outcomes is the real goal.”
Clearance helps build trust by showing that a tool works in various contexts, but she believes that successful adoption of clinical AI ultimately depends on factors like workflow integration, usability, transparency, and measurable improvement in outcomes.
She also highlighted that Bayesian intentionally took an evidence-based approach rather than a typical startup go-to-market strategy. Instead of quickly selling a first product, the company first focused on large real-world deployments and studies.
Bayesian formally spear in 2021, and the following year he had published studies In Natural medicine involving approximately 750,000 patients. Research showed that the startup’s AI demonstrated high clinician adoption rates and earlier detection of sepsis, as well as improved outcomes such as lower mortality, fewer complications and reduced length of hospital stay.
Health systems across the country are using the Bayesian sepsis tool, including Cleveland Clinic, Johns Hopkins Health System, University of Rochester Medicine And Memorial care.
One health system leader — Dr. James Leo, chief medical officer of MemorialCare’s Physician Society — emphasized that the implementation of the Bayesian tool was not just a technology deployment.
“This was an opportunity to engage our front-line staff – nurses and providers – to rethink the early identification and treatment of sepsis across our system,” he said.
MemorialCare conducted two phases in parallel. One was the technical construction of its EHR, and the other was the clinical work to ensure sepsis workflows optimize patient outcomes. Dr. Leo said the Bayesian team spoke with staff members from MemorialCare’s emergency departments, inpatient units, intensive care units and quality team to map opportunities, then the startup and health system co-designed workflows that clinicians approved before going live.
Dr. Leo highlighted that a focus on workflow design and clinician buy-in has been key to the tool’s success.
“Sepsis can progress quickly and present in subtle ways, and a patient’s risk follows them wherever they are in the hospital. Every affected nurse, provider, APP and resident is provided on the platform, which means hundreds or even thousands of clinicians working from the same information and coordinating care as a single team,” he noted.
Since adopting the tool, MemorialCare is detecting more sepsis patients earlier, with “more than double the sensitivity of what we used before,” Dr. Leo added.
He also pointed out that clinicians are working with far fewer electronic alerts, which he said helps restore their confidence in these types of AI tools. He says this reduction in alert fatigue has contributed to the platform’s high engagement rates among providers.
“When providers engage with the Bayesian flag, we see an absolute reduction in mortality of 3.6%, and the time to antibiotic administration is cut in half when they engage within the first hour. With 90% adoption in our emergency departments, our clinicians believe in it too – and that’s why we are confident in applying this principle system-wide,” said Dr. Leo.
MemorialCare’s results underscore Saria’s assertion that AI clinical tools must earn clinicians’ trust – not just regulatory approval – to truly improve patient outcomes.
Photo: Ruslanas Baranauskas/Scientific Photo Library
